He’d always been a charmer, a smooth talker. John had always known how to get what he wanted. He stood across the office, watching the congratulatory handshakes and back-slaps, his stomach churning.

His hands were fists in his pockets, nails digging into his palms. He could feel the pulse hammering in his temples, a frantic drumbeat against his skull.

He forced a smile as his boss clapped him on the shoulder, offering hollow words of praise. His voice sounded strained, unnatural, even to his own ears. He kept picturing John’s smirk.

The fluorescent lights above felt like they were shrinking the room, closing in on him. He felt trapped, suffocated. He muttered an excuse and headed to the men's room, needing a place to simply *be*.

Emotion: mad

Cluster: Anger / Hostility
PC1 (Valence): -1.94 Negative
PC2 (Disposition): 1.26

Role in Research

This story is one of 1,000 stories generated for the emotion mad. During extraction, it was fed through Gemma4-31B and its hidden state activations were captured at 11 layers.

The mean activation across all 1,000 mad stories, after denoising with neutral dialogue baselines, produces the mad emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the mad vector is projected through the unembedding matrix.

Promoted:
C0.448
0.344
est0.340
🤬0.336
aggravated0.332
Suppressed:
a-0.495
de-0.485
la-0.387
H-0.344
latter-0.290